Finding scales

Finding a scale

  • In addition to an instrument you develop yourself, your final projects require you to include two other constructs that come from established literature.

  • Finding a validated instrument relevant to your topic may required a little detective work.

  • At a minimum, you want to find a citation for your scale, the question(s) used, and the scoring method.

Finding a scale

Your best bet is probably going to be searching journal articles in the relevant field. Some good general search engines for academic research are:

Accessing library resources

  • The UMD library provides access to a ton of scholarly sources. You can access these freely any time you’re on campus.

  • If you’re not on campus, you can use the UMD reload button to redirect your request through the library proxy.

Using a search engine

Using a search engine

Using a search engine

  • This section lists the journal where the article was published. You should look for peer-reviewed sources.
  • Journal impact factor is an (imperfect) method of measuring the prestige of an academic journal using a weighted measure of citations.

Using a search engine

  • The abstract section will briefly summarize the article and the most important findings.

Using a search engine

  • There are some additional keywords here that might be helpful guides.
  • For this project, you should keep an eye out for terms like “questionnaire development”, “validation” etc.

Using a search engine

  • Web of science also links papers together using their citation networks, so this is a useful thread to pull if you want to find more literature on the same topic.

Using a search engine

  • Since I’m using the UMD proxy, a Find@UMD link shows up that will take me to a page where I can find the full text of the article.
  • In this case, the full text is freely available from the publisher.

Examining the paper

  • Academic articles tend to have a relatively consistent structure. You don’t need to read the whole thing to get useful information.

  • For our purposes here, the “methods” and “literature review” sections are probably the most important components.

Examining the paper

  • In the “methods” for this paper, there’s a heading called “measures” where the authors describe how they went about measure the key variables for their model.

Examining the paper

Examining the paper

  • The critical component is the citation for the Bergen Social Media Addiction Scale.

Examining the paper

  • The critical component is the citation for the Bergen Social Media Addiction Scale.

  • This is the original citation for the scale they’re using, so you would want to track down this article for yourself.

Examining the paper

  • If the questions are available, you’ll often find them listed in the appendix at the end, or in the supplementary materials, or with a link to a webpage.

Examining the paper

  • If the questions are available, you’ll often find them listed in the appendix at the end, or in the supplementary materials, or with a link to a webpage.

  • In this article, the questions used are listed in the appendix.

Questions to ask

  • Is the scale widely used? The citation count for the original publication can be a good indicator.

Questions to ask: versions

  • Are there different versions of the scale that might be more relevant for your project? (shorter versions, updates, minor modifications for similar constructs)

Questions to ask: validation studies

  • Have there been additional validation studies?

  • Widely used scales might have been subjected to multiple validation studies to examine how well they work and what they measure

Questions to ask: validation studies

Questions to ask: validation studies

  • A meta-analysis is a study that systematically reviews and aggregates the results of previous studies.

Questions to ask: validation studies

  • Dimensionality refers to the estimated number of latent constructs measured by the instrument (this is typically measured with some form of factor analysis)
  • Unidimensionality indicates that there’s evidence that a single trait explains most of the variation in responses to the scale items.

Questions to ask: validation studies

  • Evidence of construct validity comes from examining how well this measure correlates with other things that should be associated with social media addiction.

  • (The abstract says these are “pooled” because they’re based on a weighted average of multiple studies.)

Questions to ask: validation studies

  • Test-retest reliability is an indication of how closely responses correlate between follow-ups.

  • (There weren’t enough papers to quantitatively evaluate this)

Questions to ask: validation studies

  • Social media addiction is not a recognized psychiatric disorder, so this isn’t a clinical diagnostic tool, but it appears good enough to use to draw conclusions about problematic social media use in the population.

Questions

  • What scale or scale(s) are available?

  • Are the items accessible? Are there multiple versions? How many questions and how are they scored?

  • Are there additional validation studies? Can you find discussions about the reliability or dimensionality of the measure?

  • Is the scale widely used? Are there important criticisms or limitations mentioned? Is there a debate that warrants consideration? Competing measures?